Make getBluetoothManualChooserEvents() asynchronous.

Make getBluetoothManualChooserEvents() asynchronous.

This fixes the Windows tests, where synchronous, unpumped messages to the UI thread are forbidden: https://codereview.chromium.org/1304353004/

BUG=500989
Committed: https://crrev.com/a5cef8fe03eba361467c63ab38db5d759fa6eedb
Cr-Commit-Position: refs/heads/master@{#350167}

[Jeffrey Yasskin, 2015-09-18 20:51:33]

jyasskin@chromium.org changed reviewers:
+ mkwst@chromium.org

[Jeffrey Yasskin, 2015-09-18 20:51:33]

Please forward this to anyone else who might want to object to the nested
message loop.

I could instead make the message async by teaching
https://code.google.com/p/chromium/codesearch/#chromium/src/components/test_r...
to return a Promise, but because that's in a component and gin has no Promise
support (??), it would have to use V8's promise API directly instead of getting
to use anything like ScriptPromiseResolver.

[Mike West, 2015-09-19 09:46:37]

mkwst@chromium.org changed reviewers:
+ jochen@chromium.org

[Mike West, 2015-09-19 09:46:38]

This seems like a last-resort sort of patch.

+jochen, who will probably have suggestions.

[jochen (gone - plz use gerrit), 2015-09-21 15:08:09]

On 2015/09/19 at 09:46:38, mkwst wrote:
> This seems like a last-resort sort of patch.
> 
> +jochen, who will probably have suggestions.

the usual pattern is to give the method a callback and invoke the callback once
the async IPC came back. see any method that takes a v8::Function as parameter

[Jeffrey Yasskin, 2015-09-21 21:44:57]

The CQ bit was checked by jyasskin@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-21 21:45:42]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1351393002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1351393002/20001

[commit-bot: I haz the power, 2015-09-21 22:00:40]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-21 22:00:41]

Dry run: Try jobs failed on following builders:
  android_compile_dbg on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_compile...)

[Jeffrey Yasskin, 2015-09-21 22:04:35]

Patchset #3 (id:40001) has been deleted

[Jeffrey Yasskin, 2015-09-21 22:05:15]

The CQ bit was checked by jyasskin@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-21 22:07:28]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1351393002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1351393002/60001

[commit-bot: I haz the power, 2015-09-21 22:26:51]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-21 22:26:51]

Dry run: Try jobs failed on following builders:
  android_chromium_gn_compile_rel on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_chromiu...)

[Jeffrey Yasskin, 2015-09-21 23:01:04]

The CQ bit was checked by jyasskin@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-21 23:01:41]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1351393002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1351393002/100001

[commit-bot: I haz the power, 2015-09-21 23:28:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-21 23:28:55]

Dry run: Try jobs failed on following builders:
  android_chromium_gn_compile_rel on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_chromiu...)

[Jeffrey Yasskin, 2015-09-22 00:09:39]

On 2015/09/21 15:08:09, jochen wrote:
> On 2015/09/19 at 09:46:38, mkwst wrote:
> > This seems like a last-resort sort of patch.
> > 
> > +jochen, who will probably have suggestions.
> 
> the usual pattern is to give the method a callback and invoke the callback
once
> the async IPC came back. see any method that takes a v8::Function as parameter

'k, patch 4 is the callback-based function. Since that involved using the v8 API
directly anyway, I continued to a Promise-based API in patch 6. Let me know if I
should roll back to the callback.

The test's in https://codereview.chromium.org/1304353004/.

[jochen (gone - plz use gerrit), 2015-09-22 15:18:54]

i like the callback based version better. no messing around with global handles

[Jeffrey Yasskin, 2015-09-22 15:28:02]

That's an odd objection. Everything in the blink bindings layer uses
Persistents too, through
https://code.google.com/p/chromium/codesearch/#chromium/src/third_party/WebKi....

On Tue, Sep 22, 2015 at 8:18 AM,  <jochen@chromium.org> wrote:
> i like the callback based version better. no messing around with global
> handles
>
> https://codereview.chromium.org/1351393002/

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[jochen (gone - plz use gerrit), 2015-09-22 15:29:26]

On 2015/09/22 at 15:28:02, jyasskin wrote:
> That's an odd objection. Everything in the blink bindings layer uses
> Persistents too, through
>
https://code.google.com/p/chromium/codesearch/#chromium/src/third_party/WebKi....
> 
> On Tue, Sep 22, 2015 at 8:18 AM,  <jochen@chromium.org> wrote:
> > i like the callback based version better. no messing around with global
> > handles
> >
> > https://codereview.chromium.org/1351393002/
> 
> To unsubscribe from this group and stop receiving emails from it, send an
email to chromium-reviews+unsubscribe@chromium.org.
> 

yeah, and it's a major source of memory leaks. so if we can avoid adding more
instances and use existing patterns, that'd be great

[Jeffrey Yasskin, 2015-09-22 15:32:21]

On 2015/09/22 15:29:26, jochen wrote:
> On 2015/09/22 at 15:28:02, jyasskin wrote:
> > That's an odd objection. Everything in the blink bindings layer uses
> > Persistents too, through
> >
>
https://code.google.com/p/chromium/codesearch/#chromium/src/third_party/WebKi....
> > 
> > On Tue, Sep 22, 2015 at 8:18 AM,  <mailto:jochen@chromium.org> wrote:
> > > i like the callback based version better. no messing around with global
> > > handles
> > >
> > > https://codereview.chromium.org/1351393002/
> > 
> > To unsubscribe from this group and stop receiving emails from it, send an
> email to mailto:chromium-reviews+unsubscribe@chromium.org.
> > 
> 
> yeah, and it's a major source of memory leaks. so if we can avoid adding more
> instances and use existing patterns, that'd be great

Done.

[Jeffrey Yasskin, 2015-09-22 15:33:03]

The CQ bit was checked by jyasskin@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-22 15:35:25]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1351393002/140001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1351393002/140001

[Jeffrey Yasskin, 2015-09-22 15:46:36]

On 2015/09/22 15:32:21, Jeffrey Yasskin wrote:
> On 2015/09/22 15:29:26, jochen wrote:
> > On 2015/09/22 at 15:28:02, jyasskin wrote:
> > > That's an odd objection. Everything in the blink bindings layer uses
> > > Persistents too, through
> > >
> >
>
https://code.google.com/p/chromium/codesearch/#chromium/src/third_party/WebKi....
> > > 
> > > On Tue, Sep 22, 2015 at 8:18 AM,  <mailto:jochen@chromium.org> wrote:
> > > > i like the callback based version better. no messing around with global
> > > > handles
> > > >
> > > > https://codereview.chromium.org/1351393002/
> > > 
> > > To unsubscribe from this group and stop receiving emails from it, send an
> > email to mailto:chromium-reviews+unsubscribe@chromium.org.
> > > 
> > 
> > yeah, and it's a major source of memory leaks. so if we can avoid adding
more
> > instances and use existing patterns, that'd be great
> 
> Done.

Heh, for that matter, InvokeCallbackTask also stores several v8::Globals, and
passes them around in exactly the way the Promise-using code would. But I just
want to unblock my tests, so I'm happy to keep the callbacks version that's
currently the latest patchset.

[jochen (gone - plz use gerrit), 2015-09-22 15:49:22]

lgtm

[commit-bot: I haz the power, 2015-09-22 16:14:17]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-22 16:14:17]

Dry run: This issue passed the CQ dry run.

[Jeffrey Yasskin, 2015-09-22 16:14:41]

The CQ bit was checked by jyasskin@chromium.org

[commit-bot: I haz the power, 2015-09-22 16:15:02]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1351393002/140001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1351393002/140001

[commit-bot: I haz the power, 2015-09-22 16:20:37]

Committed patchset #7 (id:140001)

[commit-bot: I haz the power, 2015-09-22 16:21:13]

Patchset 7 (id:??) landed as
https://crrev.com/a5cef8fe03eba361467c63ab38db5d759fa6eedb
Cr-Commit-Position: refs/heads/master@{#350167}